The Hidden Cost of Turnover: How Rebalancing and Reconstitution Erode Factor-ETF Returns

Photo by Steve A Johnson on Unsplash

Every factor ETF publishes an expense ratio, and most investors treat that number as the total cost of ownership. It is not. A second cost lives inside the fund's returns: the friction generated when the underlying index rebalances its weights and reconstitutes its membership, and the fund has to trade to match. That cost is never invoiced. It is simply absorbed into net asset value, where it quietly widens the gap between what the index reports and what the shareholder actually earns.

The question worth answering is not whether this cost exists — it plainly does — but how large it is, what drives it, and whether it should change how a long-horizon investor chooses between otherwise similar factor strategies. This piece is about the mechanics of that cost and how to measure it honestly.

Context: what turnover actually is

A market-cap index like the S&P 500 is close to self-maintaining. When a stock rises, its weight rises automatically; the index rarely needs to trade to stay accurate. A factor index is different by construction. It selects and weights holdings by a rule — momentum, quality, low volatility, value, size — and those characteristics drift constantly. A momentum name that led last quarter may fail the screen this quarter. A quality name that re-rated may breach a valuation cap. To keep the portfolio faithful to its stated rule, the index periodically drops names, adds names, and resets weights. That event is reconstitution; the smaller weight adjustments in between are rebalancing.

The fund tracking that index has no choice but to transact accordingly. Turnover — the fraction of the portfolio replaced over a year — is the visible symptom. Broad market funds often run single-digit annual turnover. Rules-based factor funds can run materially higher, and the more aggressive the factor (momentum being the classic case), the higher it climbs. Turnover itself is not the cost; it is the volume across which trading friction is applied. The relationship between how factor characteristics wander over time and how often a fund must trade to chase them is something I explored separately in How Factor Loadings Drift: Watching MTUM, QUAL, and SIZE Over Five Years.

The cost decomposition

The yfinance pull for this article returned no fund-level fields, so rather than attach numbers I cannot source I want to be precise about the components instead. Trading friction on reconstitution breaks into four parts, and they behave differently:

The first two are intuitive and familiar to anyone who has traded an illiquid security. The third is the one most retail write-ups miss, and it is where the more interesting analysis lives.

The anticipation cost is the non-obvious one

A transparent, rules-based index is a public schedule of forced trades. If everyone knows that a widely followed momentum index reconstitutes on a fixed date, and the selection rule is mechanical enough to forecast, then the additions and deletions can be estimated in advance. Traders position ahead of the fund: they buy the names the index will add and sell the names it will drop, before the fund itself transacts. When the fund finally trades on the reconstitution date, it does so at prices that have already moved against it. The fund buys a little higher and sells a little lower than it otherwise would.

This is not a cost the fund manager can fully eliminate, because it is a direct consequence of the very transparency that makes the product trustworthy. The academic literature on index reconstitution has documented these predictable pre- and post-event price patterns for decades. The design tension is real: more transparent rules build investor confidence but widen the door for anticipation; more discretion narrows that door but reintroduces the manager risk that rules-based investing was meant to remove.

A transparent factor index is a public calendar of forced trades — and the shareholder pays a small toll every time the market reads that calendar before the fund does.

Initially I assumed this anticipation toll would be roughly constant across similar funds. It is not. It scales with how mechanical and how well-telegraphed the rule is, and — critically — with how much the fund needs to trade relative to the liquidity of the names involved. Which leads to the effect that matters most for anyone choosing a large, popular factor product.

Scale changes the arithmetic

Here is the asymmetry that headline turnover hides. Turnover is a percentage; market impact is not. Two funds tracking the same index can report identical turnover and yet incur very different reconstitution costs, because a $30 billion fund pushing a mid-cap name on reconstitution day moves that price far more than a $500 million fund making the proportionally identical trade. The larger fund's own demand is a bigger share of available liquidity, so its market impact per dollar traded is higher. Success — growing AUM — can therefore raise the very cost that erodes the returns responsible for attracting that AUM in the first place.

This is a genuine second-order effect: the drag is not a fixed property of the strategy, it is a function of the fund's size interacting with the liquidity of its target universe. A large-cap quality fund can absorb enormous flows with minimal impact, because its holdings trade in deep markets. A small-cap or deep-value factor fund at the same AUM faces a much steeper impact curve, because its target names are thinner. When comparing two factor funds, the AUM-to-underlying-liquidity ratio tells you more about hidden reconstitution cost than the expense ratio does. It is also why AUM and holdings overlap deserve scrutiny in their own right, a theme I took up in How Much ETF Overlap Is Too Much?

How to actually measure it

Turnover tells you the volume of trading; it does not tell you the cost of that trading. The honest measurement tool is tracking difference — the realized gap between the fund's total return and its index's total return over a defined window — combined with the fund's tax-cost ratio for taxable accounts. If a fund's tracking difference is persistently and materially wider than its expense ratio alone would explain, reconstitution friction is a leading suspect. If two comparable funds charge the same fee but one lags its index more, the difference is being paid somewhere, and trading friction is usually where.

For a long-horizon holder the practical stance is measured, not alarmed. Reconstitution drag is real, it compounds, and basis points matter over decades — but it is typically a modest annual figure that is dwarfed by asset-allocation and behavior-in-stress decisions. The point is not to fear factor ETFs; it is to compare them on realized tracking difference rather than on the sticker fee, and to remember that the cheapest expense ratio is not automatically the cheapest fund to own. The broader case for holding through these frictions rather than trading around them is one I made in Buy and Hold in 2026.

The editor holds broad factor-tilted ETFs as satellite positions and does not hold any single-factor momentum fund at the time of writing.

Frequently asked questions

Is turnover cost already included in the expense ratio? No. The expense ratio covers management and operational fees. Trading friction from reconstitution — spreads, market impact, anticipation — is absorbed into the fund's net asset value and appears as tracking difference, not as a separate charge.

Does higher turnover always mean a worse fund? Not necessarily. Turnover is the volume across which friction is applied, not the friction itself. A high-turnover fund trading deeply liquid large-caps can incur less real cost than a lower-turnover fund trading illiquid names. Measure the outcome — tracking difference — not the input.

Why does a larger fund sometimes cost more to own? Because market impact scales with size relative to the liquidity of the underlying holdings. A large fund's own reconstitution trades move prices against it more than a small fund's proportionally identical trades, so growing AUM can raise hidden trading cost even when turnover and fees are unchanged.

What is the anticipation or front-running cost? Because rules-based index reconstitution dates and rules are public and often forecastable, some traders position ahead of the fund's required trades. The fund then transacts at prices that have already moved, buying slightly higher and selling slightly lower than it otherwise would.

How do I see this cost as an investor? Compare the fund's total return to its stated index's total return over several years (tracking difference), and check the tax-cost ratio in a taxable account. A persistent gap wider than the expense ratio alone is a signal that trading friction — among other factors — is at work.

Key takeaways

• Factor ETFs carry a reconstitution and rebalancing cost that the expense ratio does not disclose; it surfaces as tracking difference.
• The cost has four parts — spread, market impact, anticipation, and realized tax — and they scale differently.
• Anticipation cost is a direct consequence of index transparency: public, dated, forecastable rebalances invite trading ahead of the fund.
• Market impact scales with AUM relative to underlying liquidity, so a fund's success can raise its own hidden cost — especially in thin small-cap or deep-value universes.
• Compare factor funds on realized tracking difference and tax-cost ratio, not on headline turnover or the sticker fee alone.

What this analysis can and can't tell you

The data pull for this article returned no fund-level fields, so it deliberately avoids fund-specific performance figures and stays at the level of mechanism. Macro anchors are sourced to FRED as of mid-July 2026 (10-year Treasury 4.58%, VIX 16.5, CPI 3.7% year over year) and describe the regime backdrop, not any fund's results. Reconstitution cost is regime-dependent — it tends to rise when liquidity thins and volatility spikes — so any single-window estimate understates the tail. To size the effect for a specific fund, pull that fund's multi-year tracking difference and tax-cost ratio directly from issuer and Morningstar data.

Methodology. Macro figures: FRED, as-of dates 2026-06 through 2026-07-14. No fund-level price or fundamentals data was available for this article; cost components are described qualitatively rather than estimated numerically, to avoid unsourced figures. Framework grounded in the published literature on index reconstitution and rebalancing.

This article is for educational purposes and does not constitute personalized financial advice. See our full Disclaimer.